HWK 13 - the two are different with commercial A being...

This preview shows page 1. Sign up to view the full content.

Homework #12 Problem #35 Page 439 Business Issue: TV Comercial messages seen and recalled by viewers. Business Question: Data: Commercial A Commercial B # who Saw 150 200 # who Remembered 63 60 Analysis: 1) Ho: Ha: 2) and std err = stdev (p-bar1 - p-bar2) p-bar A = 0.42 nA = 150 p-bar B = 0.3 nB = 200 p-bar = (n1*p-bar1 + n2*p-bar2) / (n1 + n2) = 0.35 StdError= 0.05 3) Alpha = .05 4) Norminv (.025, 0,.035)= -0.1 > pbar 0.1 < pbar 5) p-bar1 - p-bar2 = 0.12 p-value = 0.99 = 2*(1-.999665) 0.02 6) Reject Ho Business Answer According to the samples of the two commercials there is sufficent or strong evidence to be able to conclude that the recall proportions are different. The sample recall proportions of commercial A are different enough from those of commercial B to safely be able to make the conclusion that
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: the two are different with commercial A being higher. Is there any difference between the recall proportions for the two commercials. If so, what is the difference? Ho: p 1- p 2 = D =0 Ha: p 1- p 2 <> D =0 (p-bar 1- p-bar 2 ), the difference found in the two sample proportions Because when Ho is true (p 1 = p 2 ) and the sample sizes are large (n*p & n*(1-p) for both >=5) this test stat is NORMAL with mean (p-bar1 - pbar2) = D = 0 (the difference stated in Ho). = [ p-bar 1 *(1-p-bar 1 )/n 1 + p-bar 2 *(1-p-bar 2 )/n 2 ]^.5 Reject Ho if (p-bar 1- p-bar 2 ) < "value of the normal distribution with cum prob = .025" or if (p-hat 1- p-hat 2 ) > "value of the normal distribution with cum prob = .975"...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online